Coking Sour Water: A Four-Stage Physical Train Instead of Chemical Demulsification

Few refinery streams combine as many hostile properties in one pipe as coking sour water. It is acidic, it is frequently emulsified beyond what a skimmer can break, it carries enough solids to plug equipment, and it dissolves organics that biology would rather not touch. Treating it with the generic oily-water toolkit is how plants end up with a sludge disposal problem on top of a water problem.

What makes coking sour water different

The stream forms when refineries process high-sulphur crude oil and its by-products - coke, coal gas and tar. It dissolves a large amount of acidic components, hydrogen sulphide and sulphuric acid among them, and carries sulphides, ammonia, phenols, cyanides and oils together. Composition goes well beyond that short list: phenols, oils, naphthalene, pyridine, quinoline and anthracene, along with other heterocyclic and polycyclic aromatic compounds, are all present. It is widely regarded as one of the harder problems in industrial wastewater treatment, and the description is fair.

Four difficulties, taken one at a time

• Acidity. A high concentration of acidic substances and a low pH make the stream difficult to handle, and the acids are acutely toxic to the environment and to living organisms, so they cannot be left in the effluent.
• Emulsification. Some streams arrive severely emulsified, and conventional methods struggle to achieve demulsification at all, let alone oil removal.
• Solids and oil together. Large amounts of suspended solids and oily contamination make the water turbid and harder to treat, and the combination blocks and damages equipment as well as depressing performance.
• Organics and heavy metals. Various organic substances and heavy metal ions are harmful to ecosystems, so removal has to cover them alongside the acids and the solids.

Why the familiar answers fall short

Each of the conventional answers below has a place in a larger scheme. What none of them provides is a way to break an emulsion without adding something to the water, and that is the gap the physical route addresses first.

Activated sludge is the standard aerobic answer and it does reduce oil, some organics, ammonia nitrogen and total nitrogen to a degree. Coking wastewater, however, carries large amounts of refractory macromolecular organics that conventional activated sludge cannot remove completely, so the effluent keeps a high organic content and fails both discharge and reuse standards. Advanced treatment is then needed anyway.

Chemical methods neutralise the acids by dosing reagents, which works on paper and creates a solids problem in practice: large amounts of chemical sludge and oily waste residue are produced, secondary pollution follows easily, and reagent costs are relatively high. Biological methods use microbial communities to degrade organics and purify the water, and they offer good performance with little secondary pollution, but operation is complex and the growth environment of the community has to be controlled and maintained within narrow limits. None of the three deals with the emulsion at its source.

The physical train SINOKLE built

The patented process system is centred on CDFU cyclonic dissolved gas flotation, fine filters and the KHC high-efficiency coalescing de-oiler. Its defining property is that the entire treatment is purely physical demulsification and separation: no chemical reagent is added at any stage, so no scum and no oily sludge are generated and secondary pollution is avoided at the root rather than managed afterwards. Demulsification performance is stable and repeatable, and the separated waste oil is recovered in a condition good enough for reprocessing, with good oil quality and low water content. Energy saving, environmental performance and economic return arrive together, which is unusual in this duty.

What the train does not do is remove dissolved phenols and cyanides, and it is worth saying so before a project expects it. Those fractions pass through a purely physical de-oiling stage by design, which is why the downstream biological or oxidation step still has to be sized for them. The physical train solves the emulsion, the solids and the oil recovery problem; it leaves the dissolved organics question exactly where it was, and a specification that confuses the two will end in an effluent that is clear and still non-compliant.

Capacity and numbers from a petrochemical site

The CDFU unit installed at a petrochemical plant for coking sour water runs a coarse filter, CDFU cyclonic dissolved gas flotation, fine filtration and KHC high-efficiency coalescing de-oiling in series. Capacity is 15 m³/h. Inlet wastewater oil content is ≤150,000 ppm (15%), and effluent oil content is ≤50 ppm on average, with the recovered oil going back to reprocessing rather than to waste.

Two things about that train deserve attention when a similar project is being specified. The first is the fine filter: it exists to protect the KHC media, and skipping it converts a coalescer into a filter that has to be cleaned constantly. The second is the absence of a chemical package - no demulsifier, no neutralising reagent, no flocculant - which removes an entire operating cost line and an entire category of sludge from the plant's permits.